查詢結果分析
相關文獻
- Demonstration of Cognitive Modeling in Categorization: Fitting Two Neural Network Models to the Data from Yang and Lewandowsky (2003)
- Neural Network Procedures for Taguchi's Dynamic Problems
- A Fast and Efficient Competitive Learning Design Algorithm Based on Weight Vector Training in Transform Domain
- 專家系統振動訊號圖型判別之研究
- 反傳遞模糊類神經網路於流量推估之應用
- 類神經網路(Neural Networks)的種類及其在影像處理上的應用
- C++Fuzzy類神經網路物件導向發展系統之建立
- 臺灣汽保費率之估計--對數線性費率模式與類神經網路之比較
- 運用類神經網路於股價指數之套利--以日經225指數為例
- 使用類神經網路預估碳化鎢材料放電加工性能
頁籤選單縮合
題名 | Demonstration of Cognitive Modeling in Categorization: Fitting Two Neural Network Models to the Data from Yang and Lewandowsky (2003)=認知模擬在類別學習上的應用:以Yang與Lewandowsky (2003)之研究為例 |
---|---|
作者 | 楊立行; Yang, Lee-xieng; |
期刊 | 中華心理學刊 |
出版日期 | 20070900 |
卷期 | 49:3 2007.09[民96.09] |
頁次 | 頁285-300 |
分類號 | 176.3 |
語文 | eng |
關鍵詞 | 認知模擬; 類別學習; 類神經網路; Cognitive modeling; Categorization; Neural network; |
中文摘要 | 為了探索人類心理歷程與心智表徵,各種不同的認知模型不斷地被研究者提出。這些認知模型代表著不同的理論觀點,它們不僅可以對現象提出解釋,還可以對未知進行預測。當我們想要檢視一個理論模型對於現象的可以達到多好的解釋力,以實徵資料進行電腦模擬就成為了一項強而有力的研究工具。然而本地(台灣)的心理學背景的學生往往缺少學習這項工具的管道,而不清楚什麼是電腦模擬、不知道如何進行,不了解它的重要性何在。這部分可能源自於心理系所鮮少開設相關的課程,也或者是它需要較高的程式設計能力。因此,本文目的在於提供進行認知模擬的概念性引導方針:文中將首先介紹兩個在類別學習領域上相當知名的類神經網路模型ALCOVE與ATRIUM,並以Yang與Lewandosky(2003)知識分化的研究為例進行電腦模擬,透過此二模型與實徵資料的分別比對結果,進一步對此二模型背後所支持的理論進行比較分析。模擬結果顯示,ATRIUM對於在類別學習上的知識分化現象的解釋力明顯高於ALCOVE。此外,一些相關的理論層次的議題,如分類表徵的異質性等,也因為獲得模擬的結果而能夠被更深入的討論。 |
英文摘要 | In investigating human mental processes and mental representations, a cognitive model represents a theoretical view, provides explanations to the observed phenomena and makes predictions about an unknown future. When evaluating how well a theory can account for the phenomenon of interest, modeling is a powerful research tool. However, local (Taiwanese) psychology students have limited exposure to what cognitive modeling is, how to do implement cognitive models, and why cognitive modelling is important. This is partly due to a lack of university courses that teach cognitive modelling and partly due to the demands that modelling places on one's skills. The purpose of this article is to provide a conceptual guideline of how to do modeling, by fitting two neural network models-ALCOVE and ATRIUM to the data from the study of Yang and Lewandowsky (2003), which tested the theoretical concept of knowledge partitioning in categorization. The modeling results show that ATRIUM outperforms ALCOVE in accounting for the knowledge partitioning results. Some relevant theoretical-level discussions, such as the heterogeneity of categorization, are also included. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。